| From: | "Chad Wagner" <chad(dot)wagner(at)gmail(dot)com> |
|---|---|
| To: | "Rob Schall" <rschall(at)callone(dot)net> |
| Cc: | "Stephan Szabo" <sszabo(at)megazone(dot)bigpanda(dot)com>, pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: Identical Queries |
| Date: | 2007-03-01 21:09:17 |
| Message-ID: | 81961ff50703011309v43c0974eie31172e36f8460cd@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On 3/1/07, Rob Schall <rschall(at)callone(dot)net> wrote:
>
> There are 4 entries (wanted to make the playing field level for this
> test). There are 2 with true for istf and 2 with false.
>
Then the difference here has to do with using orignum vs destnum as the join
criteria. There must be more intersections for orignum than destnum, or
your statistics are so far out of whack. It appears to be estimating 5M vs
500K for a result set, and naturally it chose a different plan.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alex Deucher | 2007-03-01 21:11:25 | Re: strange performance regression between 7.4 and 8.1 |
| Previous Message | Alex Deucher | 2007-03-01 21:06:34 | Re: strange performance regression between 7.4 and 8.1 |